Do States with Strict Gun Laws have Significantly Different Violent Crime Rates from States with Loose Gun Laws?
We are constantly hearing on the news that we need more gun control and that it would reduce gun violence. At first glance, it seems a reasonable assumption. For example, if you restrict how many people can own bats, its likely you would reduce the number of violent incidents involving a bat.
But are we really trying to stop incidents of gun violence or are we interested in reducing violent crime in general? In other words, if reducing the number of guns available to the general public actually increased violent crime rate overall, or even had no effect on violent crime rate, would it make sense to ban guns? If we stop gun related suicides but have no impact on suicides overall, have we accomplished anything?
A lot of people will use statistics to convince you that something is good or bad without pointing to these limitations because it leads to questions that may have answers that go against what they want you to believe.
I don't want to portray myself as an unbiased researcher. On the contrary, I'd like to make sure you are aware of my biases and allow you to come to your own conclusion. In general, I do not believe gun laws are bad, when they do not punish law abiding citizens. However, I am mostly against any sort of restriction on access to semi-automatic firearms. My line of reasoning is that legal gun owners actually prevent crimes because criminals are less likely to commit a crime knowing that the person they are committing a crime against is very likely to use lethal force to defend them self. As a result, I'm interested in seeing what data indicate about the restriction of guns on crime rates.
The first thing I did was pull up a ranking of states by how strict they are in terms of gun laws.
This was taken from a website that is in favor of stricter gun laws: You can visit that website here.
Overall, you can find very misleading statements like "Many of the states with the strongest gun laws also have the lowest gun death rates nationwide."While this may be true statistically, it doesn't say much about how these gun restrictions may impact overall crime rate. Its almost equivalent to banning or restricting cars and saying "By restricting cars, we've reduced the total number of car related deaths". Its easy to see that statements like this completely disregard other effects such a ban or restrictions may have on other things (like getting to work).
So once I collected this list of top 10's, I decided to find out how many violent crimes per 100,000 people were committed in each state. I chose this ratio because it would be unfair to compare states total number of violent crimes since they all vary in total population which is a very good predictor of total number of violent crimes (see my earlier post). In order to make a fair comparison, a ratio was necessary.
Next I plugged the data into SPSS. Here is the data:
1.00 Connecticut 272.80
1.00 California 411.10
1.00 New Jersey 308.40
1.00 Massachusetts 428.40
1.00 Hawaii 287.20
1.00 New York 398.10
1.00 Maryland 494.10
1.00 Illinois 429.30
1.00 Rhode Island 247.50
1.00 Michigan 445.30
2.00 South Dakota 254.10
2.00 Arizona 405.90
2.00 Mississippi 269.80
2.00 Vermont 135.20
2.00 Louisiana 555.30
2.00 Montana 267.50
2.00 Wyoming 219.30
2.00 Kentucky 238.20
2.00 Kansas 353.90
2.00 Oklahoma 454.80
I grouped the states by Strictest (1) and Loosest (2). Next I checked to see if the data were normally distributed. This is important because if the data are not normally distributed, it violates an assumption that is necessary for doing a statistical test.
Under Sig, we can see the probability is greater than 10% for both groups, which means we can assume normality. Also, the t-tests are pretty robust even when the data are not normal. Meaning, if we violated this assumption we are only less likely to find a difference between the Strict and Loose group.
Next, I ran an independent samples t-test to analyze the data.
The data did not come out significantly different. Meaning, the number of violent crimes from a state with strict gun control was not significantly different from those that had loose gun control laws.
However, I noticed something interesting when I looked at the box-plots.
The dark line in the middle of the box is the average number of violent crimes in each group. Group 1 is the states with strict gun laws and group 2 is the states with loose gun laws. You'll notice from these plots that the average number of violent crimes is actually LOWER in group 2. Also, the little arms the stick out from the box indicate outliers (extreme values). If you look at the raw data I posted above, you'll notice there is one extreme state from group 2, Louisiana.
If Louisiana would have had an average number of violent crimes (or less), the states with loose gun laws would have actually had a significantly lower number of violent crimes. However, speculating how fudged data will affect conclusions could go the other way too.
Conclusion: Gun laws do not prevent or decrease violent crimes and may even increase them. Its important to question the statistics we are presented with because they are often misleading and frame things in a way that fail to take into account other things that should also be considered when passing legislation.
Personal opinion: Guns laws can be good. On the other hand, when you start punishing law abiding citizens for isolated incidents of gun violence, you are tossing the baby out with the bath water. Lowering crime rates and shootings requires educating the public and a conversation about mental health. Lets stop talking about guns as if they are the sole reason our society has tragic events.